String Inverters vs Microinverters vs Power Optimisers: What Actually Changes

Six years installing panels on Californian roofs taught me one thing above all else: the inverter topology argument gets more airtime online than it deserves for a simple, unshaded roof, and not nearly enough airtime for a roof with a chimney shadow crossing three panels every afternoon.

String inverters, the default for a reason

A string inverter takes DC output from a series "string" of panels and converts it to AC at one central point, usually mounted on a wall near the switchboard. It's the cheapest topology per watt, the easiest to service (one box, one location, no need to get back on the roof for most inverter faults), and for a roof with a single orientation and no shading, it performs essentially identically to the alternatives. The catch is that a string is only as strong as its weakest panel. If one panel underperforms, whether from shading, dirt or a fault, the whole string's output drops toward that panel's level, because the panels in a string operate at the same current.

String inverters have also gotten genuinely better in the last several years. Multiple maximum power point tracking (MPPT) inputs on a single inverter let a system split into two or three separate strings, each tracked independently, which softens some of the weakest-link problem without the full cost of per-panel electronics. For a lot of straightforward roofs this middle step already solves most of the practical issue people worry about.

Microinverters, one converter per panel

Microinverters put a small inverter behind each individual panel, converting DC to AC right there rather than sending DC down to a central unit. Each panel then operates independently, so shading or underperformance on one panel doesn't drag down its neighbours. You also get per-panel production data, which is genuinely useful for spotting a fault early. I've caught failing panels within days using per-panel monitoring that would have taken months to notice as a small dip in whole-string output.

The tradeoff is cost and serviceability. You're buying as many inverters as panels instead of one central unit, and a failed microinverter means a roof visit to swap it out, versus a ground-level swap for a string inverter fault. In six years of installs I saw string inverter faults more often, just because there were more string systems out there, but the per-unit failure rate of microinverters wasn't meaningfully different from a well-specified string inverter in my experience. Longevity claims on both sides get thrown around a lot in sales conversations; I'd ask to see the actual warranty term in writing rather than take either claim at face value.

Power optimisers, a middle path

Optimisers sit between the two: a small DC-to-DC converter behind each panel that conditions the panel's output before sending it, still as DC, to a central string inverter. You get most of microinverters' per-panel shading tolerance and monitoring, at a lower cost than full microinverters, because the central inverter is still doing the heavier DC-to-AC conversion work. It's a genuinely sensible option for a roof with partial shading on some panels but not others, where full microinverters would be overkill and a plain string inverter would leave real production on the table.

Where this actually matters: shading and multi-orientation roofs

If your roof is a single clean plane with nothing casting a shadow on it, ever, the topology choice is close to a wash and I'd default to the cheaper string inverter option. Where it matters is a roof with a tree, a neighbour's second storey, an antenna, or, very commonly in established suburbs, a chimney that throws a moving shadow across part of the array for part of the day. On those roofs, string inverters lose disproportionately more output than the shaded panel's own share would suggest, because of how series strings behave. Microinverters or optimisers recover most of that loss by letting each panel do its own thing.

Multi-orientation roofs, east-facing and west-facing arrays on the same system, common on hip roofs, are the other case where per-panel or per-string independence pays off, letting each orientation's array produce to its own potential rather than being matched to the weaker-producing side through a shared string inverter's maximum power point tracking.

Monitoring, and why it matters more than people expect

Whatever topology you choose, ask what monitoring comes standard versus as a paid add-on. Whole-system monitoring tells you the array is producing roughly what's expected in aggregate, but it can hide a single underperforming panel behind an otherwise healthy-looking total. Per-panel monitoring, which comes naturally with microinverters and optimisers, and is available as an add-on for some string inverter systems, is the difference between catching a fault within a week and only noticing it eighteen months later when a quarterly bill looks unusually high. I push almost every client toward some form of panel-level visibility now, even on a clean roof, simply because it makes faults visible early rather than invisible for years.

What I'd actually specify

For a clean, single-orientation, unshaded roof, I still put clients into a good string inverter, ideally with multiple MPPT inputs, and save the premium. For anything with real shading, multiple roof planes, or a client who genuinely wants per-panel monitoring to catch faults early, I move to optimisers as the sensible middle ground, and reserve full microinverters for roofs with the worst shading problems or clients who specifically want the resilience of fully independent panels. I'd argue the industry oversells microinverters as a universal upgrade. On a genuinely clean roof you're paying for shading tolerance you'll never need.

Whichever topology you land on, ask your installer to show you the actual string design or panel layout for your specific roof, not a generic brochure comparison. That's where the real answer to "which one do I need" lives. See our guide to sizing a home solar system for the broader sizing conversation this sits inside, and our hybrid inverter listing for a concrete example of a battery-ready hybrid unit. If shading is your specific problem, our piece on how shading wrecks solar payback goes deeper on quantifying it before you buy.

For independent technical background on inverter topologies and certification, the US National Renewable Energy Laboratory's solar research publishes genuinely technical comparisons, and in Australia the Clean Energy Council's approved products list is the place to check that any inverter you're quoted is actually certified for use here.

Lifespan and replacement timing

One thing that gets missed in the topology debate is that inverters, of any kind, don't last as long as panels. A string inverter typically carries a shorter warranty than the 25-year panel warranty you'll see quoted, and most homeowners will replace at least one central inverter over a system's lifetime. Microinverters spread that replacement risk across many smaller units rather than one central point of failure, which some clients value even beyond the shading argument, since a single microinverter failure knocks out one panel's production rather than the whole array while you wait for a technician. Whichever way you go, ask what the inverter warranty actually covers, parts only or parts and labour, and for how long, since that gap matters more than people expect once a unit is a decade old and out of its original warranty window.

— Sam Whitfield, Solar & batteries